This is the website of Abulsme Noibatno Itramne (also known as Sam Minter).
Posts here are rare these days. For current stuff, follow me on Mastodon
|
OK, now I know that I will never, ever, ever, get to the bottom of that list of 500 DVDs. But still, I like adding things whenever I think of them or hear about them. You suck. Its not like you have to actually send me all of those DVDs, you just have to keep them in a silly list for me. And you stop me at 500? Bleh.
Perhaps I need to start keeping my own backlog for video, just like I do for books and music. Bleh.
I’ve been trying to keep a little dashboard showing how well Backblaze is doing keeping up with backing up my primary system (a Mac). I’d been doing this by hand every once in awhile by just taking the numbers of files and MB left shown by their icon in the menu bar. But that was manual and tedious. I wanted to automate it. I Googled a number of times things like “Backblaze Command Line” to try to find a way to pull this info, but never found anything useful. I poked around a bit looking for relevant log files, but didn’t find them until tonight. I found what I needed, so I thought I’d post so that if anybody else is looking to do the same thing, when they Google, they will find the answer, rather than having to dig and find this from scratch. Of course this is just for the Macintosh version. I am sure there is something equivalent in their Windows software, but I don’t have a Windows machine, so I’m not going to be the one to find it there.
Anyway…
It turns out what you need is all in a nice little file:
/Library/Backblaze/bzdata/bzreports/bzstat_remainingbackup.xml
The contents are pretty straight forward:
<?xml version=”1.0″ encoding=”UTF-8″ ?>
<contents>
<remaining remainingnumfilesforbackup=”2414327″ remainingnumbytesforbackup=”459544254971″ />
</contents>
So, for remaining files needing to be backed up:
cat /Library/Backblaze/bzdata/bzreports/bzstat_remainingbackup.xml | grep remaining | cut -d \” -f 2
And for remaining bytes needing to be backed up:
cat /Library/Backblaze/bzdata/bzreports/bzstat_remainingbackup.xml | grep remaining | cut -d \” -f 4
I use these together in a quick little shell script that I run every few hours using cron:
#!/bin/bash
# Updates BackBlaze Stats
FCFILE=/Users/abulsme/Dev/MailCount/bbfiles.txt
MBFILE=/Users/abulsme/Dev/MailCount/bbmeg.txt
date “+%m/%d/%Y %H:%M:%S” | tr -d ‘\n’ >> $FCFILE
echo -ne ‘\t’ >> $FCFILE
cat /Library/Backblaze/bzdata/bzreports/bzstat_remainingbackup.xml | grep remaining | cut -d \” -f 2 >> $FCFILE
date “+%m/%d/%Y %H:%M:%S” | tr -d ‘\n’ >> $MBFILE
echo -ne ‘\t’ >> $MBFILE
cat /Library/Backblaze/bzdata/bzreports/bzstat_remainingbackup.xml | grep remaining | cut -d \” -f 4 >> $MBFILE
My actual script actually does one more thing to convert bytes to megabytes for my charts, but the above gets you the raw data. What you do with it after that depends on your application and what you are trying to do with this info.
For me, I’m generating dashboard charts:
Everything prior to 2010 Jun 10 was me grabbing data manually for the graph. June 10th forward is pulled automatically by the above.
Anyway, hope this ends up being helpful for anybody else wanting to programmatically monitor how well Backblaze is or is not keeping up.
Edit 2012 Jul 9 13:02 UTC: In recent versions of Backblaze, the path that contains the report has changed to /Library/Backblaze.bzpkg/bzdata/bzreports/ .
Giving up for tonight. Hopefully what I have now should work for Facebook but not Twitter if tonight’s pattern holds. And it doesn’t hit LinkedIn like ping.fm did. But ping.fm was messing up the links (via Superfeedr) so whatever. It is late. I may revisit this again someday when I have more time. For now, it is what it is.
That last way of doing things worked for Facebook but not Twitter. Looks like lots of people are having a similar issue with Twitterfeed recently. Trying something else. Bleh.
I’ve still got open slots in my team at Amazon. Yes, rather that make a little puzzle like my last post I’m saying it specifically. :-) We have a lot of fun and challenging things we are working on, and Amazon in general is a great place to work. Full updated job description for my open positions:
Automated and Targeted Merchandising
The Automated and Targeted Merchandising team delivers relevant merchandising content to users who are most likely to respond favorably to that content. Our team manages a wide range of automated, highly scalable systems, for delivering targeted email and website content to receptive customers. We also own a number of systems that assist marketing and merchandising professionals to find, schedule, and apply relevant targeting for any subset of users. Automated and Targeted Merchandising systems rely heavily on computational advertising, data mining and machine learning techniques. These techniques, combined with aggregated customer behavior data, allow us to provide a high degree of accuracy in targeting the right demographics – and driving higher sales as a result.
Qualifications
The ideal candidate will have strong development skills in Java or C++, with experience in designing and architecting scalable high-performance systems and services that are also flexible, reliable, and maintainable. They will have excellent problem-solving skills, a solid understanding of computer science fundamentals, a quantitative mindset along with a proven track record of on-time delivery. We are seeking engineers that are highly pragmatic and can solve real-world problems faster than others believe possible. Candidates should be familiar with machine learning, data analysis and data mining techniques. Database skills and experience with large-scale multi-tiered distributed systems are highly desired. Familiarity and comfort with front end web technologies a plus. Advanced degree in Computer Science with a strong Mathematical background preferred.
If you are a Software Developer and the job description above sounds like it might be up your alley, toss me a resume at abulsme@abulsme.com. If that specific position isn’t quite what you might be interested in, the Website Application Platform group I am a part of has a bunch of other spots open as well, not just software engineers. See the WAP Jobs site for more information.
If you don’t fit any of the above yourself, but know someone who does, send them my way too. :-)
Last month I posted that my company was hiring. They still are. But now I am also hiring for my own group. If you are a Software Developer (and I know some of you are) and you might be able to be convinced to come to the Seattle area for a good position at a good company on a good team doing interesting things, please toss me a resume at abulsme@abulsme.com and I’ll give you some more background and get you into our system to start the process of looking to see if you would be a fit for the spots I am hiring for, or other places within the company. The specific job description for my team:
[My] team manages a number of systems which enable [us] to target highly relevant content to the users most likely to respond positively. Our systems span the range from completely automated email and website programs to systems which assist human schedulers in finding and applying relevant targeting. Our systems rely heavily on computational advertising, data mining and machine learning techniques. [Our] team is looking for truly exceptional software engineers with strong technical abilities and a passion for this area.
The ideal candidate will have strong development skills in Java or C++, with experience in designing and architecting scalable high-performance systems and services that are also flexible, reliable, and maintainable. They will have excellent problem-solving skills, a solid understanding of computer science fundamentals, a quantitative mindset along with a proven track record of on-time delivery. We are seeking engineers that are highly pragmatic and can solve real-world problems faster than others believe possible. Candidates should be familiar with machine learning, data analysis and data mining techniques. Database skills and experience with large-scale multi-tiered distributed systems are also highly desired. Advanced degree in Computer Science with a strong Mathematical background preferred, minimum 3-5 years work experience required.
Note that we are looking for engineers with less experience as well. So if the 3-5 years listed at the end of the above is the only thing holding you back, email me anyway.
And if you don’t meet the description above, but know someone who does, send them my way too.
As with my post last month, I’ll mention that I make a habit of not naming my company explicitly online (at least on my blog, you can find it at LinkedIn and elsewhere if you look and pay attention), but it is a Fortune 500 company based in the Seattle area which is not Costco, Microsoft, Paccar, Weyerhaeuser, Starbucks, Nordstrom or Expeditors, and it is in the tech/retail space and is named after a geographic feature in South America… if you can’t figure it out with that, we don’t want to hire you anyway. :-)
After initially setting up some stuff with twitterfeed, I decided to try ping.fm instead. At the same time, I upgraded this Postie thing thing that I use to sometimes post via email. Between those two things my last post went kind of nuts and didn’t do what I wanted. So more testing here. Apologies in advance. Regular programing will resume shortly.
For quite awhile now, maybe a year, maybe less, dunno, but for awhile, I have had the Facebook “Notes” app set to import the RSS feed from my blog. So basically each of my posts would show up as a Note in facebook. This seemed to work pretty well. My posts would usually show up on Facebook an hour or two after I posted them on my blog. And I started to get more comments to my posts via Facebook than directly on my blog. Which is cool I guess.
But lately, the last month or so, the delay between me posting and things showing up on Facebook was getting longer and longer. A few hours turned into half a day. That turned into about a day. Then it was several days. Then a little over a week ago I noticed that none of my posts had moved over in almost a week. So I went in to the Notes application and reentered the address of my RSS feed. It responded by quickly importing several of my recent posts it hadn’t gotten yet.
But then I made more posts. It has once again been a few days (not quite a week yet) and nothing has come through. So I think I am done with that method.
I’ve been using TwitterFeed for quite a few months to dump my blogging into Twitter. Turns out it can do the same for Facebook. (I could also add a level of indirection and go from Twitter to Facebook, but…) Anyway, I set it up a little bit ago, and see it already posted my most recent blog post.
Unlike the notes, it is just a headline and a link to the post on my blog, rather than a full copy of the whole blog post. Guess that means less people will actually read any of them, but I guess that’ll be OK. If you are one of the folks who looks at my posts only through Facebook, you’ll just have to click through. Or just go to Abulsme.com directly of course.
For now, I haven’t actually turned off the notes feed. In theory it is still supposed to work. I’m curious if it will ever start posting stuff again. Maybe it will, just with a two week delay or some such. We’ll see. If it does start doing something again, I’ll probably turn off one or another method of doing this to avoid duplicates depending on how things are going at the time.
Of course, I know there is also the idea that just having the same things posted to my blog, Facebook and Twitter defeats the purpose of them as separate platforms, and maybe I should just use each of them in distinct ways. Maybe. But as far as I’m concerned, my blog, where I have full control of what is in it, I have the backups, I could switch hosting whenever I want, etc, etc, etc is my single place for anything I want to put online. Everywhere else can have pointers back, but I’m not going to put stuff I really care about in someone else’s system as the primary place it lives. Pointers linking back are just fine. It is bad enough that I am using Flickr and YouTube for some pictures and videos. But the original copies of all that content remains safely in my care. I just use them for convenience. :-)
Anyway, we’ll see how TwitterFeed works for this purpose. Theoretically, with this posts on my blog should always show up on Facebook (and of course Twitter) within 30 minutes. That will be better than what the Notes import was doing even when it was working, let alone when it started lagging by days… if it is even working any more.
If anybody has any thoughts on this way of doing things, or on alternative ways to accomplish this kind of thing, I’d love to hear them. Thanks!
And they look to be overwhelmingly positive so far. This is going to get interesting. Gizmodo has a good round up here.
|
|